This project will demonstrate the feasibility of identifying circulating tumor cells (CTCs) that have undergone an epithelial-mesenchymal transition (EMT) by profiling a gene expression signature using a targeted sequencing platform, the RASL-Seq assay. The presence of CTCs that have undergone EMT is prognostic of poor long term survival. The gene profile will also include genes for identifying and counting EpCAM+ and EpCAM- CTCs, and for identifying the expression of targets for therapeutic agents. Thus, the RASL-Seq CTC Profiling signature assay will deliver the EpCAM+ counts that the current Veridex CellSearch(R) assay does, but provide significant additional utility by identifying EpCAM- cells that have undergone EMT (not assessed by CellSearch) and which put the patient at risk, as well as provide the oncologist with an indication of what therapy these cells may be responsive to, overall significantly improving patient care and survival outcome. The EMT and risk signature used in this proposal is the subject of patents filed by MD Anderson. BioSpyder, which has separately licensed the RASL-Seq technology from UCSD, will demonstrate its feasibility in Phase I, determining if gene expression profiling of individual CTCs is required or if pools of (e.g. EpCAM- and EpCAM+) CTCs can be used for testing. Initial development and validation will be with cell lines, then a set of 30 patient samples, and finally a set of 200 patient samples. CellSearch results for these patients will be compared to the RASL-Seq assay results to determine the positive predictive value of the RASL-Seq test for Epcam+ CTCs. The results of this program will demonstrate that the RASL-Seq profiling assay has broad applicability not just to CTC analysis, but to analysis of biopsies, and for use as a companion diagnostic.